Hi everyone,
We’ve just released ServiceControl 6.21.1.
Fixed bugs
- #5948 Migrated Lucene indexes are reset to Corax at start-up when the index is not locked
- #5960 Retry processor runs a dynamic query that creates the auto index
Auto/RetryBatches/ByStatus
How to know if you are affected
You are affected if:
- you use ServiceControl 6.20.0 or 6.21.0 with RavenDB persistence, and
- the error or audit database was created by a version before 6.20, and
- you migrated one or more indexes from Corax to Lucene in RavenDB Studio without locking them
Symptoms
For #5948, ServiceControl resets a migrated index to Corax at the next start-up. RavenDB builds a replacement index, named ReplacementOf/<index>, with Corax and swaps it in. The migration is lost and the index is rebuilt, which can take a long time on large databases. The start-up log and the Error Database Search Engine or Audit Database Search Engine custom check report the index as using Corax again.
After upgrading, migrated indexes no longer need to be locked. If an index was already reset to Corax, follow the recovery steps in the search engine documentation.
For #5960, a minor inconvenience is fixed. The Error Database Search Engine custom check kept failing after all indexes were migrated to Lucene, because of an auto index that ServiceControl should not have created.
When to upgrade
You should upgrade immediately if you are affected. Otherwise, you should upgrade during your next maintenance window.
Where to get it
You can download the new version of ServiceControl from our website or Docker Hub.
Please read our support policy for more details.
With thanks,
The team in Particular